报告嘉宾：邹丹平 ( 上海交通大学 )
报告时间：2016年3月30日晚20:00（北京时间）
报告题目：《StructSLAM：Visual SLAM with Building Structure Lines》
主持人： 郭炜炜（上交）
报告相关文献列表： StructSLAM：Visual SLAM with Building Structure Lines, Huizhong Zhou, Danping Zou, Lin Pei, Peilin Liu, Wen Xian Yu, IEEE Trans Vehicular Technology, 2015 报告摘要：
Indoor scenes usually exhibiting strong structural regularities. In this talk, Danping will introduce a novel visual SLAM method that adopts building structure lines as novel features for localization and mapping. The advantages of the proposed method are twofold: 1) line features are easier to be detected and tracked in texture-less scenes; 2) the orientation information encoded in the structure line feature can reduce drift error significantly. Experimental results show that the proposed method performs better than existing methods in terms of stability and accuracy. The talk will present the motivation and implementation in details. 讲者简介：
Danping Zou is an assistant professor in School of Electronic Information and Eletrical Engineering, Shanghai Jiao Tong University ( SJTU ). Before 2013, He was a research fellow in Department of Electrical & Computer Engineering at National University of Singapore. He graduated from Huazhong University of Science & Technology ( HUST) in 2003 and got the PhD degree in School of Computer Science at Fudan University 2010.
His research focuses on designing real-time navigation systems for small drones, which includes the following topics: Visual SLAM, Integrated Navigation, Micro-UAVs
报告材料[Slides] | 